Why We Invested in Ragie
Making it easier for developers to incorporate AI into their apps
By David Sacks and Brian Murray
Today we are excited to announce our investment in Ragie. Craft led the $5.5 million seed round, along with participation from Valor, Chapter One, and Saga.
Nearly every team building AI-powered apps needs to incorporate Retrieval Augmented Generation (RAG). RAG ingests data from a large corpus (e.g. enterprise documents or SaaS application data), retrieves the most relevant information based on user queries, and generates contextually accurate responses or content by combining the retrieved data with generative AI capabilities.
However, developing and maintaining a bespoke RAG service is costly. The process involves connecting multiple sources, extracting data from various formats, implementing evolving chunking and retrieval techniques, building scalable data processing pipelines, and avoiding hallucinations. Even after investing a lot of time, developers find their homegrown solutions are too brittle and unreliable.
Ragie is doing for RAG infrastructure what AWS did for the cloud. It enables builders to move faster and stay current with state-of-the-art RAG methods via a maintenance-free service. It offers a robust API for ingesting and retrieving content using the latest techniques in RAG for chunking, searching, and re-ranking. Through a streamlined developer experience, developers can easily connect and sync their applications with data in Google Drive, Notion, and Confluence. Many more app connectors are coming soon.
Our journey with Ragie began through our own experience incubating our work chat app Glue. We needed an easier way to query documents shared on Glue, so the content could appear in GlueAI’s answers. Ragie helped us deliver in 3 weeks instead of 3 months, showcasing its ability to significantly accelerate development timelines.
Cofounders and seasoned industry veterans Bob Remeika and Mohammed Rafiq are the minds behind Ragie. Their expertise and vision have led to the development of advanced features such as ‘Summary Index’ to avoid document affinity problems and ‘Entity Extraction’ for extracting structured data from unstructured documents.
With its straightforward pricing model, including a free tier for developers to get started, Ragie is poised to change how teams implement RAG in their AI applications. Developers who use Ragie not only avoid the hassle of developing their own RAG infrastructure, they also get the benefits of Ragie’s constantly improving service through its simple-to-use APIs.
We look forward to seeing Ragie empower developers and businesses to harness the full potential of their data in AI applications.
